#!/usr/bin/python
CACHEDIR = '/home/rubys/planet/cache/'

import cgi, os, sys, time
print 'Content-Type: application/xml,charset=utf-8\r\n\r\n',

# read POST parameters
fs = cgi.FieldStorage()
query=fs.getvalue('query')
startTime=int(fs.getvalue('startTime') or time.time()+3600)
lastFound=int(fs.getvalue('lastFound') or time.time())
startFile=fs.getvalue('startFile')

# prime results
import libxml2
msg=[]
libxml2.registerErrorHandler(lambda msg,str: msg.append(str), msg)
results = libxml2.parseDoc('<results/>')
root = results.getRootElement()

# sniff test xpath, if invalid, proceed no further
try:
  ctxt = results.xpathNewContext()
  ctxt.xpathRegisterNs('atom','http://www.w3.org/2005/Atom')
  ctxt.xpathRegisterNs('xhtml','http://www.w3.org/1999/xhtml')
  ctxt.xpathRegisterNs('mathml','http://www.w3.org/1998/Math/MathML')
  ctxt.xpathRegisterNs('svg','http://www.w3.org/2000/svg')
  ctxt.xpathRegisterNs('xlink','http://www.w3.org/1999/xlink/')
  ctxt.xpathRegisterNs('planet','http://planet.intertwingly.net/')
  ctxt.xpathEval(query)
except:
  # TODO: add errror message to output
  print results.serialize('utf-8')
  sys.exit(1)

# sort Venus cache by date
cache = os.listdir(CACHEDIR)
cache = [(os.stat(CACHEDIR + name).st_mtime,name) for name in cache]
cache.sort()
cache.reverse()

# scan Atom entries
parsed = found = 0
start = time.time()
for mtime, file in cache:
  if os.path.isdir(CACHEDIR + file): continue

  # make sure that we resume where we left off
  if mtime > startTime: continue
  if startFile and mtime == startTime:
    if file == startFile: startFile = None
    continue
    
  entry = None
  try:
    try:
      entry = libxml2.parseFile(CACHEDIR + file)
      ctxt = entry.xpathNewContext()
      ctxt.xpathRegisterNs('atom','http://www.w3.org/2005/Atom')
      ctxt.xpathRegisterNs('xhtml','http://www.w3.org/1999/xhtml')
      ctxt.xpathRegisterNs('mathml','http://www.w3.org/1998/Math/MathML')
      ctxt.xpathRegisterNs('svg','http://www.w3.org/2000/svg')
      ctxt.xpathRegisterNs('xlink','http://www.w3.org/1999/xlink/')
      ctxt.xpathRegisterNs('planet','http://planet.intertwingly.net/')
      
      if ctxt.xpathEval(query):
        found = found + 1
        lastFound = mtime
        root.addChild(results.copyNode(entry.getRootElement(),1))
    except Exception, e:
      pass # print e
  finally:
    if entry: entry.freeDoc()
    parsed = parsed + 1

  # if we found enough or have searched for too long, quit
  if found >= 10 or time.time()-start > 0.3:
    # Policy: no bookmark if nothing found in a 90 day stretch.  Limits
    # run-away scans of the entire database for queries that match nothing
    # TODO: let users manually continue via pressing 'OK'.
    if lastFound - mtime <= 86400*90:
      # add bookmark
      bookmark = root.newChild(None,'bookmark',None)
      bookmark.setProp('query',str(query))
      bookmark.setProp('startTime',str(mtime))
      bookmark.setProp('startFile',str(file))
    break

# add stats
stats = root.newChild(None,'stats',None)
stats.setProp('scanned',str(parsed))
stats.setProp('found',str(found))

# return results
print results.serialize('utf-8')